Abstract
In repeated visual search, little is known about whether foreknowledge of a target-defining feature in an upcoming search affects a current search already. Therefore, we investigated in three experiments, to what extent oculomotor behavior during two consecutive searches is affected by the knowledge of the target color of the second search. Participants searched a display consisting of pink and blue letters twice. They were informed that, in Search 1, the target could be pink or blue, whereas in Search 2, the target color was constant. In order to measure whether foreknowledge of the target color affects both searches, we analyzed saccadic latencies to a probe presented on a relevant or irrelevant item during Search 1 and at the beginning of Search 2 (Experiment 3 only). Participants were requested to saccade to this probe as fast as possible and then to continue the search. Furthermore, we analyzed saccadic latencies and number of fixations to (unprobed) relevant and irrelevant items. Across experiments, we found, in Search 1, shorter saccadic latencies to unprobed relevant compared to irrelevant items but no differences in latencies to the probed (relevant or irrelevant) item. Also, there was no difference in the number of fixations to relevant and irrelevant items. In contrast, in Search 2, we found shorter saccadic latencies to both unprobed and probed relevant compared to irrelevant to items. Furthermore, participants inspected more relevant than irrelevant items. Taken together, oculomotor programming in repeated search is, at least to some extent, affected by a prospective target feature and can adapt immediately once a change in relevance occurs.
Meeting abstract presented at VSS 2018